Breaking Down Walls Improving Healthcare for Resilient Populations - PDF document

2/28/2019 Breaking Down Walls Improving Healthcare for Resilient Populations through Stakeholder Engagement? Tung Nguyen, MD Professor of Medicine, UCSF Medical Care of Vulnerable and Underserved Populations February 28, 2019 Disclosures  Dr. Nguyen’s spouse works for Gilead Sciences, Inc., a pharmaceutical company. No pharmaceutical products will be discussed.  Dr. Nguyen works for UCSF, an academic medical center. 1

2/28/2019 Outline  The Case for Change  Stakeholder Engagement  Examples of Stakeholder Engagement to Address Health Disparities  Next Steps We Have The Highest Healthcare Costs  2017: $3.5 trillion (17.9% of GDP)  Compared to similar countries  Per capita spending ($10,739) is twice as high  Administrative, labor, and drug costs are highest  Outpatient utilization rates are similar  Hospitalization rates and length of stay are lower 2

2/28/2019 Our Healthcare System is Worst Our Health Outcomes Are Worst Papanicolas et al, JAMA 2018 3

2/28/2019 We Have the Worst Health Disparities Income Mortality Gap • Compared to top 1% in income, the poorest 1% live 10.1 years women) or 14.6 years (men) less. • The gap has widened from 2001 to 2014. The Association Between Income and Life Expectancy in the United States, 2001-2014 JAMA. Published online April 10, 2016. doi:10.1001/jama.2016.4226 Healthcare Access Makes Little Difference among the Poor Life Expectancy, Bottom Quarter in Income • Health behaviors and where they live have a significant impact on the life expectancy of the poor. • Access to medical care is not a major determinant of life expectancy among the poor. The Association Between Income and Life Expectancy in the United States, 2001-2014 JAMA. Published online April 10, 2016. doi:10.1001/jama.2016.4226 4

2/28/2019 Doing the Same Thing and Expecting Different Results is …  More health insurance coverage will improve outcomes, but by how much and at what cost?  Are we spending too much on healthcare and not enough on other social services?  Our patients spend most of their lives outside of the healthcare system.  Does “population health management” make sense when it is being managed from within the healthcare system?  Are we doing enough with patients and communities to achieve health outside of clinic and hospital walls? Outline  The Case for Change  Stakeholder Engagement  Examples of Stakeholder Engagement to Address Health Disparities  Next Steps 5

2/28/2019 Who Are Stakeholders in the Care of Individual Patients?  Patients  Caregivers  Healthcare providers  Healthcare staff  Healthcare systems  Health insurers  ??? Who Are Stakeholders in “Population Health”?  Patients and caregivers  Healthcare providers, staff, and systems  Home health agencies, medical interpreters, etc…  Health insurers  Public health department  Patient advocacy groups  Community-based organizations  Employers  Media  Government  ??? 6

2/28/2019 Why Stakeholder Engagement?  Who should decide:  What diseases matter or problems that should be prioritized?  What are appropriate processes of care?  What health outcomes matter?  How should these processes and outcomes be achieved?  How are corrective measures determined and applied?  What level of evidence should be used to make decisions and priorities? Definition and Levels of Stakeholder Engagement for Health  The involvement of key members and partners of the group(s) affected by the issue that builds on their interests and strengths, and combines knowledge with intervention to improve health.  Level of engagement: Advisory, Steering, Executive  Stages of engagement:  Prioritization  Design  Implementation  Assessment  Dissemination  Sustainability 7

2/28/2019 The Science of Stakeholder Engagement  Patient-Centered Outcomes Research (PCOR)  Practice-Based Research  Community-Based Participatory Research (CBPR) Outline  The Case for Change  Stakeholder Engagement  Examples of Stakeholder Engagement to Address Health Disparities  Next Steps 8

2/28/2019 An Example of Patient-Centered Outcomes Research to Address Health Disparities Health Within Reach: Using Mobile Technology to Improve Asian American Health Patient-Centered Outcomes Research Institute (PCORI) AD-12-11-4615 Overview • Asian Americans have low screening rates for hepatitis B and C and high rates of liver cancer. • Goal: use patient-centered outcomes research to create tech to address health • Develop, implement and measure effect of English, Cantonese, Mandarin, and Vietnamese iPad-based interactive app to increase hepatitis B and C screening among Asian American patients. 9

2/28/2019 Stakeholder Engagement • Prioritization of topic, identifying intervention, grant writing: SF Hep B Free (community organization), AANCART (community network), Hep B QIC (systems network, DPH) • Implementation: SF Hep B Free • Oversight: Patient Advisory Councils , AANCART, Hep B QIC • Focus groups and interviews: community members and patients, clinic staff, physicians, medical directors . • Patient Advisory Councils: barriers and responses, application look and feel (buttons, fonts, colors, flow) video look and feel, languages, control materials, pilot test • Patients: pilot test of application Application Look and Feel 10

2/28/2019 Mobile Application: Questions Mobile App and Provider Alert 11

2/28/2019 Pilot Testing Cluster Randomized Trial • Randomization: Primary Care Physician (PCP) • Hepatitis: 70 PCPs and patients • Nutrition & Physical Activity: 52 PCPs and patients • PCPs in both arms received list of patients not screened for hepatitis B every 6 months (Provider Panel Notification) • Prior to PCP visit, both group of patients: • Used assigned mobile app. • Received bilingual printout summarizing the assigned topics and tailored recommendations (Provider Alert) and asked to give to PCP. 12

2/28/2019 Patient Eligibility • Asian American • Age 18+ • Speak English, Cantonese, Mandarin, or Vietnamese • Clinic patient (visit within 3 years) • No hepatitis B screening (surface antigen) test in electronic record. Hepatitis B Discussion: 3 months E ve r disc usse d he patitis B with 70.4 doc tor 16.5 Doc tor e ve r r e c omme nde d 51.1 te sting for he patitis B 13.2 E ve r aske d doc tor to ge t 58.9 he patitis B te st 9.3 0 10 20 30 40 50 60 70 80 All p <0.001 He patitis NPA 13

2/28/2019 Intervention Effects: Multivariate Models Intervention 95% Confidence Odds Ratio Interval Hepatitis B Test 7.1 3.5, 14.1 ordered within 3 months Hepatitis B Test done 7.3 3.5, 15.1 within 3 months Hepatitis C Test done 9.7 5.6, 16.7 within 3 months (birth cohort) Intention to treat Not Needy But Resilient 14

2/28/2019 An Example of Community-Based Participatory Research to Address Health Disparities Lay Health Worker Intervention to Promote Colorectal Cancer Screening Among Chinese Americans Funder: National Cancer Institute Stakeholder Engagement  Prioritization of topic, identifying intervention, grant writing: NICOS Chinese Health Coalition , AANCART  Intervention development and pilot: focus groups of community members and leaders  Implementation: NICOS Chinese Health Coalition, lay health workers  Oversight: AANCART  Dissemination: NICOS Chinese Health Coalition 15

2/28/2019 Dual Cluster Randomized Controlled Trial Randomizes 58 lay health workers (LHWs) into 29 Experimental LHWs 29 Comparison LHWs Recruit 360 experimental participants Recruit 365 comparison participants Pre-educational session survey Pre-educational session survey Two LHW sessions on Two health educator lectures on CRC screening + healthy eating & physical activities + CRC brochure CRC brochure Post-educational session survey Post-educational session survey Chinese Colorectal Cancer Screening Flipchart 16

2/28/2019 Lay Health Workers Characteristics of Chinese American Participants (N=725) Sociodemographics % Male 19% Married 74% Limited English proficiency 95% Less than high school education 72% Income < $20,000 60% Health and health care access Fair/ Poor 65% Has at least 1 chronic health condition 60% Visited MD in the last 12 months 80% Has regular place of care 90% Uninsured 9% 17

2/28/2019 Multivariable Models of Intervention Effects Ever Had CRC Up-to-date for CRC Screening Screening Intervention Effect 1.94 (1.34, 2.79) 2.02 (1.40, 2.90) US Residence >10 yrs 1.65 (1.11, 2.46) 1.37 (0.94, 2.00) Fair/poor health 1.52 (1.07, 2.15) 1.29 (0.97, 1.73) Had regular place for healthcare 1.81 (1.01, 3.25) 1.81 (0.99, 3.29) Had primary care doctor 2.64 (1.42, 4.92) 2.66 (1.47, 4.83) Have health insurance 2.51 (1.34, 4.68) 2.60 (1.37, 4.94) Model adjusted for LHW cluster, age, gender, education, income, marital status, English fluency, employment Capacity Building 18